Data analyst moving to Fabric or Databricks
A data analyst moving to Microsoft Fabric should take PL-300 if reports and semantic models are the daily work and then DP-600 once the job includes lakehouses, warehouses and SQL; an analyst moving to Databricks should take the Databricks Data Analyst Associate. The engineering exams (DP-700, DP-750, Databricks Data Engineer Associate) are the right answer only if the next job is building pipelines rather than querying and presenting data.
Example profiles, not real people. The scenarios below are illustrative planning examples built from the official exam outline. They are not reviews or reports of anyone’s result.
The decision rule
Certify the platform your next job runs, at the level of the work you will do there. An analyst’s work falls into three layers: querying and shaping data, modelling it for others, and presenting it. Each exam below covers a different mix of those layers.
| Your next role | Exam to take | Why |
|---|---|---|
| Report builder in a Power BI shop | PL-300 | Power Query, modelling, DAX, report design and the Power BI service |
| Owner of semantic models and Fabric data stores | DP-600 | Lakehouses, warehouses, SQL, KQL, DAX and governance |
| Analyst on Databricks SQL | Data Analyst Associate | Queries, dashboards, query analysis and Genie spaces |
| Pipeline builder on Fabric | DP-700 | Ingestion, transformation, streaming and monitoring |
| Pipeline builder on Azure Databricks | DP-750 | Environment, Unity Catalog, data processing and deployment |
| Pipeline builder on Databricks itself | Data Engineer Associate | Ingestion, transformation and Lakeflow Jobs |
If two rows fit, choose the one that matches the tool you will open on your first Monday.
The Fabric route: PL-300 first, then DP-600
PL-300 is the analyst exam. It runs 100 minutes at a passing score of 700 out of 1000 and splits its weight almost evenly: prepare the data, model the data and visualise and analyse the data at 25 to 30 percent each, with managing and securing Power BI at 15 to 20 percent. For an analyst who already builds reports, much of it describes daily work.
DP-600 is a different job. Its largest domain, Prepare data, carries 45 to 50 percent and covers choosing between data stores, transforming data with views, functions and stored procedures, and querying with SQL, KQL and DAX. Semantic models and maintaining the solution carry 25 to 30 percent each. There is no report-design domain at all, so strong visuals earn nothing there. The PL-300 vs DP-600 comparison lays the two outlines side by side, and the DP-600 prerequisites page gives a self-check for SQL, DAX and KQL.
You do not need PL-300 before DP-600; Microsoft lists no prerequisite. The sensible order is a matter of where your gaps are. An analyst whose DAX is shaky benefits from PL-300 first because the modelling skills carry directly into DP-600’s semantic model domain.
The English DP-600 outline changes on October 19, 2026, so check which outline applies to your booking date.
The Databricks route: Data Analyst Associate
The Data Analyst Associate is 45 scored questions in 90 minutes for a $200 registration fee, English only, valid for two years. Databricks does not publish a passing score. Nine sections, none above 20 percent, mean the exam rewards coverage: executing queries 20, dashboards 16 and analysing queries 15 percent form the analyst loop that makes up just over half, and Genie spaces adds 12 percent. Importing data and data modelling are 5 percent each, so this is not a pipeline exam.
Genie spaces is the section most analysts have not touched. At 12 percent it is worth about five questions, which is more expensive than a familiar area at a higher weight. The analyst vs data engineer comparison shows how the two Databricks associate exams divide the work.
When an engineering exam is the honest answer
If your real plan is to stop answering questions and start building the supply of data, the analyst exams will feel light. Take DP-700 for Fabric pipelines and Real-Time Intelligence, DP-750 for Azure Databricks, or the Databricks Data Engineer Associate. DP-750 vs DP-700 compares the two Microsoft exams, and the Databricks vs Azure data engineering guide helps when the platform is still undecided. The site’s own advice is blunt: a certification in a platform you do not touch is worth very little in an interview, so let the employer’s platform decide.
Realistic constraints
Time. The PL-300 and DP-600 study plans both assume four weeks at 8 to 10 hours a week for someone who already works with the tools. The Databricks analyst plan is five weeks at 6 to 8 hours. Without hands-on access to a Fabric workspace or a Databricks SQL warehouse, expect to need longer, because both are practical exams.
Language. The Databricks Data Analyst Associate is English only. PL-300 is offered in ten languages and DP-600 in seven, with Microsoft saying localised exams follow an English outline update by roughly eight weeks. See the English as a second language guide.
Renewal. Microsoft associate credentials last 12 months and renew through a free online assessment, so Fabric and Power BI certifications are a yearly habit. Databricks credentials last two years and renew only by sitting the full current exam again at the same fee. Holding both routes means two different maintenance rhythms.
Retakes. Microsoft allows a retake after 24 hours and then 14 days between later attempts. Databricks asks for 14 days between attempts with no cap, and every attempt is paid. The retake guide compares the rules.
Scenario: an analyst who writes SQL daily and builds Power BI reports
This is an illustrative planning example. The analyst writes SQL every day, publishes reports to the Power BI service and has never created a lakehouse. The goal is a role that owns Fabric semantic models.
Where the time goes: almost half the exam sits in Prepare data, which is the part the analyst has touched least in a store-level sense. The trap is to spend the first weeks polishing DAX, which is familiar, and discover too late that choosing between a lakehouse, a warehouse and an eventhouse is a recurring question type.
A reasonable planning assumption is the DP-600 plan’s four weeks at 8 to 10 hours, 32 to 40 hours in total, with roughly half of those hours on Prepare data to mirror its 45 to 50 percent weight, and the remainder split between semantic models and governance. This is a planning figure, not a promise of any result. If PL-300 is not yet held, the same analyst could take it first, since its modelling content is the foundation for DP-600’s semantic model domain.
Where to go next
Read the Microsoft data and analytics path if the choice is among Microsoft exams, or the Databricks certification path if the platform is Databricks. Then open the DP-600 study plan or the Data Analyst Associate study plan and run the free practice test in the exam section you chose.